Nicholas Catasso 7

Journal of Universal Language 12-1

March 2011, 7-46

The Grammaticalization of
Demonstratives:
A Comparative Analysis

Nicholas Catasso
Università Ca’ Foscari di Venezia

Abstract
An article, irrespective of its distribution across natural languages,
dialects and varieties, is a member of the class of determiners which
particularizes a noun according to language-specific principles of
grammatical and semantic structuring. Definite articles in Indo-
European languages – in those grammatical systems where they are
present – are derived from ancient demonstratives through a
grammaticalization process: given that demonstratives are deictic
expressions (i.e. they depend on a frame of reference which is
external to that of the speaker and of the interlocutor) with the role
of selecting a referent or a set of referents, it is easy to understand
what the role of “universal quantifier” of the, which is in English the
prototypical – but questionable – example of definiteness, is due to.
Demonstratives are frequently reanalyzed across languages as
grammatical markers (very often as definite articles, but also as
copulas, relative and third person pronouns, sentence connectives,

Nicholas Catasso
Dipartimento di Scienze del Linguaggio - Dorsoduro 3462 - VENEZIA 30123 (VE)
Phone 0039-3463157243; Email: nicholas_catasso@libero.it
Received Oct. 2010; Reviewed Dec. 2010; Revised version received Jan. 2011.
8 The Grammaticalization of Demonstratives: A Comparative Analysis

focus markers, etc.). In this article I concentrate on the

grammaticalization of the definite article in English, adopting a
comparative-contrastive approach (including a wide range of Indo-
European languages), given the complexity of the article.

Keywords: grammaticalization, definite articles, English, Indo-

European languages, definiteness

1. Introduction

In this paper, I investigate the incidence of the grammaticalization

of demonstratives into articles from a comparative perspective,
arguing that a correlation between the diachronic development of
deictic expressions and that of the pragmatic-semantic concept of
definiteness is definitely present. I took into account not only the
typological – and therefore statistical – tendency of demonstratives
to lose a [+demonstrative] (Lyons 1999) or [+deictic] (Giusti 2001)
feature and become definite noun determiners (whatever may be
intended as “definiteness”) in the majority of Indo-European
languages, but also a number of considerations which can be drawn
from the observation of languages in synchrony.
First of all, in order to understand this process, which I argue is
ultimately attributable to the context factor1, we need to categorically
1
According to Traugott’s definition, grammaticalization in general can be considered
as “the process whereby lexical material in highly constrained pragmatic and
morphosyntactic contexts is assigned grammatical function and, once grammatical,
is assigned increasingly grammatical, operator-like function”. Meillet (1912)
anticipated this concept of pragmatics – and its decisive nature over the whole
process – by introducing the importance of “expressivity” (which he intended
more or less as semantics, pragmatics) in the domain of grammaticalization. He
notes, for example, that in Latin the role of word order was “expressive”, not
grammatical. A sentence can be rendered through different word orders: Catullus
amat Lesbiam, Lesbiam amat Catullus, Amat Catullus Lesbiam, Amat Lesbiam
Catullus, and so forth. The fixing of word order has a grammatical function and
bears at least two of the main features of grammaticalization: firstly, it does not
 Nicholas Catasso 9

discriminate (purely) grammatical versus lexical items. This

distinction is basic in morphology for the semantic categorization of
morphemes bearing either lexical or grammatical meaning. According
to their function, open word classes, i.e. lexical items, normally
include adjectives, verbs (excluding auxiliary verbs), nouns,
prepositions2 – used to report or describe qualities and actions; on
the other hand, closed word classes – indicating relationships and
definiteness/indefiniteness, linking parts of the discourse, etc. – are
composed of finite sets of words and are more reluctant to the
addition of new members, although they may undergo phonetic
modifications over long periods of time. Generally speaking, though,
a form can be said to be “grammaticalized” when, having a content
word as a starting point, it develops and assumes the grammatical
features of a function word. Considering that in most languages not
all function words are phonologically and syntactically independent,
it is possible to discriminate grammatical forms that display a more
labile bonding with other grammatical units from forms which
imply a less loose relationship (e.g., affixes, which are attached to a
stem to form a new word). In particular Hammond and Noonan
1988 in Theoretical Morphology: Approaches in Modern Linguistics
speak of a “continuum of bonding” characterized by a set of
variability from one form to another. The focal areas are, in order of
bonding:

a) Prepositions bearing locative meaning with a full

segmental and prosodic structure such as in This is the
city I would love to live in, which show to have no
reductions of any kind and are fully stressed (in this

modify existent grammatical tools, but creates new ones; in the second place, it
implies a development from expressive to grammatical meaning.
2
The status of prepositions as contentives or as functors is problematic in current
syntactic theory (for an exhaustive discussion of the topic and its variants see
Cruttenden/Faber 1991)
10 The Grammaticalization of Demonstratives: A Comparative Analysis

case, the full preposition is absolutely independent);

b) Derivational morphemes adding a meaning component
(which are, therefore, neither inflections nor clitics) and
affecting or not the category in question 3 . Classical
examples are the prefix re–, indicating repetition, and
the suffix –er, deriving a noun from a verb and in
particular designating persons from objects of their
labour or occupation. Since derivational affixes are
bound morphemes, they basically differ from compounds
in that free morphemes are used in that case to form
new words;
c) Clitics, which differ from affixes in a number of
linguistic features (though they must appear next to their
host, i.e. an autonomous word), as Anderson (2006: 33)
notes: they have a lower degree of selection with respect
to their hosts; they are less likely to have idiosyncratic
shapes and semantics as compared to affixed units;
differently from affixes, they can be attached to material
already containing clitics. From a certain point of view,
clitics may be considered as being halfway between
autonomous words and affixes, as they share particular
features, e.g. the fact that they form an accentual unit
with the host;
d) Inflections are endings that carry grammatical information
such as tense and number (e.g. the French singular/
plural contrast la maison – les maisons).

In the second place, it is of fundamental importance to determine

the nature of definiteness. Lyons (1997) argues that what we call

3
While suffixes frequently affect the category of the item (for instance, Eng. -ly
derives adverbs from adjectives), prefixes rarely do: for example en- forms verbs
from nouns and adjectives (enlarge, encircle, etc.).
 Nicholas Catasso 11

“definite articles” need not be associated with a particular

[+definite] feature: definiteness is not necessarily realized in lexical
form (for instance by a definite item) but it is rather grammaticalized
structurally through the specifier of the functional projection D,
whatever may occupy this position in any given occasion. Although
I support Lyons’ basic assumption that definite articles are not
necessarily the only source of definiteness in the NP structure, I
argue here that NPs bear a feature which could be defined as “force
of the NP” (henceforth FNP), occupying a higher functional
projection and determining the grade of definiteness of the nominal
expression on the basis of the context. Arguing for an application of
the “Force” concept to the NP structure, I refer to the notion
introduced by Rizzi (1997) in his analysis of the left periphery of the
English sentence, i.e. the conclusion that the C system involves a
structure of the type FORCE – (TOP*) – FOC – (TOP*) – FIN – IP.
Given the structure of the split CP, I argue for a symmetrization of
CP and NP, assuming the same deep representation. I propose that
NPs are inherently definite (one consideration which is generally
related to proper names) and this “[+definite] feature” is licensed or
not by the context. Definite articles – and other determiners – are
generated in the functional projection D and move then to FNP to
receive the definite feature they are interpreted with. I agree with
Giusti (1997: 103-105) that the definite article has no semantic
content but only encodes structural case, i.e. it corresponds to the
morphological case which Indo-European languages have been
losing in the course of their diachronic development. What is more,
for the Balkan languages displaying a postponed determiner (e.g.
Romanian, Macedonian, Bulgarian) the same structure can be
assumed:

Eng. [FNP[FN'[FN° thei[DP[D'[D° ti[NP[N'[N° flowers]]]]]]]]]

Rom. [FNP floare-jFN'[FN° -lei[DP[D'[D° ti[NP[N'[N° tj]]]]]]]]]
12 The Grammaticalization of Demonstratives: A Comparative Analysis

In the absence of a context which allows a definite interpretation

for the NP, the phrase remains undefined (i.e. articleless in the
languages which display an article system) or receives an indefinite
interpretation (*She ate pear vs. She ate the pear vs. She ate a pear).

2. The Question of the Origin of the

Category “Article”

An article – regardless of its distribution across languages,

varieties or dialects – is a member of the class of determiners that
restricts or particularizes a noun indicating the type of reference
made by it, with very peculiar uses which in usage are to be
considered language-specific.
In my analysis I will focus on the origin and development of
definite articles in Indo-European languages – in those grammatical
systems where they are present – which derive from ancient
demonstratives. Considering that demonstratives are deictic expressions
(i.e. they depend on a frame of reference which is external to that of
the speaker and the interlocutor) which serve to select a referent or a
set of referents, it is easy to understand what the character of
universal quantifier of such an article as “the”, which is the
prototypical example of definiteness in English, is due to. As
pointed out by Diessel (2008: 12), demonstratives are frequently
reanalyzed across languages as grammatical markers (very often as
definite articles, as in the case considered here, but also as copulas,
relative and third person pronouns, focus markers, sentence
connectives, etc.). I will concentrate in particular on English and
German as representatives of the Germanic language family
preferring, nevertheless, a comparative and contrastive perspective
on the topic, given the complexity and variability of the issue.
The question of the origin of demonstratives has been the subject
of much interest in the grammaticalization research, although it has
 Nicholas Catasso 13

not been entirely resolved. Demonstratives are generally considered

to be grammatical items and the grammaticalization theory claims
that all grammatical development involves shifts in specific linguistic
contexts from a [+lexical] to a [+grammatical] item4, but there is “no
evidence from any language that demonstratives derive from a
lexical source or any other source, for that matter, that is non-
deictic” (Diessel 1999: 44; also cf. Himmelmann 1997: 21). Traugott
(1982: 245) says, for example: “It is dubious whether we can trace
all grammatical markers derived by processes of grammaticalization
to lexical items rather than to certain seemingly fundamental
grammatical items, such as demonstratives pronouns and interrogatives.
The Indo-European t- demonstrative and kU-interrogative, for
example, have been remarkably resistant to change over several
thousand years, and no lexical source seems recontructable for them”.
Next to the classical interpretations of the definite article as a
determiner with two main functions (that of particularizing or of
specifying as in “The apple that I have eaten” and that of
generalizing as in “The dog is the man’s best friend”), there are a
number of interesting views, starting from Lyon’s (1999: 290), who
defines the function of articles as similar to that of expletive subjects.
For Giusti (1997: 103-105) the article has no semantic content (like
definiteness, for instance), but only encodes structural case. In this
sense, she analyzes the functional projection containing NPs “not as
a Determiner Phrase (DP), but rather as a Functional Projection (FP),
whose head F° is reserved for nominal case and articles”.
It is interesting to notice that, even though in Indo-European no
traces of definite or indefinite article can be found and most ancient
Indo-European languages lack it either, it is considered to be a
typical late Indo-European feature, which appeared when the
languages of this family started existing separately. In the same way

4
For an exhaustive explanation of this unilateral tendency also cfr. Haspelmath’s
1999 article Why is Grammaticalization Irreversible?
14 The Grammaticalization of Demonstratives: A Comparative Analysis

Homer’s (as well as Mycenaean) Greek had no article, but this

category was already widely in use in all classical Greek dialects.
This is also the case in Romance languages, although Latin had no
article at all.
A general assumption of grammaticalization is that the
diachronic development of definite articles is not independent of its
context (i.e. of the syntactic structure it occurs in). While Lyons
(1999) argues for the loss of a [+dem] feature, Giusti (2001)
hypothesizes that one of the most outstanding differences between
demonstratives and definite articles is the presence vs. absence of a
[+deictic] feature, although both accounts crucially relate this
development to the loss of a feature.5 It is compelling to consider the
following hypothesis for the reanalysis of demonstratives as definite
articles: instead of arguing that “there is a dependency formed
between the demonstrative, which occurs in a lower functional
position in the nominal structure, and the D head” (Roberts &
Roussou 2003: 136) and that this relation, which is “mediated via
movement” (cf. supra), is exclusive, I propose that this diachronic
interconnection should expand to comprehend a FNP node: the
merger of the lexical item (i.e. the demonstrative) in D is followed
by a Move > Merge reanalysis in FN, in which the contextual
adequacy is licensed or not licensed. As we can see in (a.) and (b.),
the contextual adequacy constraint has idiosyncratic results from a
semantic point of view – considering an unmarked context in which
the NP has a generic value:

a. I like FNP[ø DP[ t NP[Ballett]]]

b. Amo FNP[la DP[ t NP[danza]]]
c. Îmi place FNP[dans-ul DP[NP[t]]]

5
In particular, Lyons argues that the grammaticalization of the distal pronoun ILLE
in Lat. is due to the earlier restructuring of the pronominal system, whereas Giusti
relates it to the loss of morphological case marking.
 Nicholas Catasso 15

For Romanian, in which the article occupies a different position,

we can hypothesize that the enclitic article moves from D° to FN°,
whereas the NP dans moves to [Spec, FNP] through [Spec, DP].

2.1. Back to the Roots: the Demonstrative in PIE and in

Proto-Germanic

The history of demonstrative expressions in all modern Indo-

European languages is extremely difficult to reconstruct. It ideally
finds its origins in PIE which, as a reconstructed language,
represents the abstract (methodological) limit of historical linguistics.
According to the historical-comparative approach, it has been
reconstructed as a language displaying among other things a
complex case system. The early emergence of this particular class of
linguistic expressions depends on the fact that their function is very
basic in human communication 6 : as Diessel (1999a: 2) notes,
demonstratives orient the hearer in the speech situation, focusing his
or her attention.7 The changes generally occurring with the gramma-
ticalization of demonstratives are of two types: functional, in the
sense that the grammatical items evolving from demonstratives are no
longer used to focus the interlocutor’s attention on entities in the
outside world and they are usually deictically non-contrastive;
syntactic, because logically their occurrence is often restricted to a
specific syntactic context and they are obligatory to form a given
grammatical structure (Diessel 1999: 2).
Beekes (1995) reconstructed a demonstrative system in Indo-
European with only two pronouns: *so (meaning both this and that)
and *h₁ anaphoric element indicating the “just named”). PIE, as a

6
Because of the basic meaning of demonstratives, it is not a case that this grammatical
category constitutes one of the first to be learnt by children in L1 acquisition
(Diessel, 1999a: 110). In fact, Lat. demonstrare = show, indicate.
7
Also cfr. http://web.unirsm.sm/DCom/2003/Functional/Abstract/Diessel%20handout
%20San%20Marino.pdf
16 The Grammaticalization of Demonstratives: A Comparative Analysis

non-attested language, is naturally very difficult to reconstruct and

this is particularly true for demonstrative pronouns, 8 in particular
because, given the many forms of demonstratives among Indo-
European languages, the common method of comparison between
these languages produces an unrealistic number of stems which
make the framework confused. Here is the system proposed by
Beekes, in turn based on Lane’s theory (1961: 469):9

Table 1. Proximal-distal Demonstratives in Indo-European

Demonstrative Pronouns (Beekes)
Singular Plural
Masculine Neuter Feminine Masculine Neuter Feminine
Nominative *so *seh2 *toi *seh2i??
*tod *teh2
Accusative *tóm *teh2m *tons *teh2ns
Genitive *(to)sio *(t)eseh2s *tesom? *tesom?
Ablative *tosmōd *toios?
Dative *tosmōi *tesieh2ei *toimus *teh2mus?
Locative *tosmi *tesieh2i *toisu *teh2su?
h
Instrumental *toi? *toi? *toib i *teh2bhi?

Tracing the development of this grammatical category in the history

of the English language starting from its roots in Proto- Germanic10,

8
Fortson (2004), for example, reconstructed * h1 as *ei-. We will adopt Beekes’ view
here.
9
Lane elaborated a mechanism of binding a certain number of standard particles to
a basic stem. The best-known example of this regards the development of the
word this.
10
Although the reconstruction of Proto-Germanic is difficult because of the absence
of written material, researchers assume, observing the development of the attested
languages, that by 250 BC Proto-Germanic had already branched into five groups
of Germanic.
 Nicholas Catasso 17

we know that this hypothetical proto-language had a demonstrative

which could serve as both a demonstrative adjective and a
demonstrative pronoun.

Table 2. Anaphoric Demonstratives in Indo-European

Nominative *h1e *(h1)ih2 *h1ei *ih2es
*(h1)id *ih2
Accusative *im *ih2m *ins *ih2ns
Genitive *h1éso *h1eseh2s? *h1es(om)
Ablative *h1esmōd *h1eios?
Dative *h1esmōi *h1esieh2ei *h1eimus
Locative *h1esmi *h1esieh2i *h1eisu
Instrumental *h1ei? *h1eibhi

Ringe (2006) has given the following paradigm, where we notice

that the same s- is present in the masculine and feminine nominative
singular, whereas all other forms have a þ-, as in Old English:

Table 3. Demonstratives in Proto-Germanic

Masculine Feminine Neuter
Singular Plural Singular Plural Singular Plural
Nominative *sa *þai *sō
*þō,
*þen(ō), *þōz *þat
Accusative *þans *þō *þiō
*þan(ō)
Genitive *þes(a) *þezō *þezōz *þaizō – –
*þesmō, *þemiz,
Dative *þezai *þaimiz – –
*þasmō *þaimiz
Instrumental *þiō – – – – –
Locative *þī – – – – –
18 The Grammaticalization of Demonstratives: A Comparative Analysis

2.2. The Development in Old English and Middle English

In Old English (approximately 7th century-110011), which displays

two types of demonstratives – the weak and the strong demonstrative
pronouns –, definiteness could still be expressed through the weak
and strong form of the adjective (i.e. without articles). Nevertheless,
as Hogg and Denison (2008: 117) argue, some kind of article
already occurred in OE prose, in association with the weak form of
the adjective, to indicate definiteness. For instance, in the following
example, the definite NPs are preceded by a demonstrative in
combination with the weak form of an adjective – unspedigan, rican
(Hogg & Denison 2008: 117, from The Sermones Catholici, or
Homilies of AElfric12) :

(1) Caseras he geceas ac þeah he geendebyrde þone unspe-

digan fiscere ætforan þam rican casere
‘emperors he chose and yet he ranked the unwealthy
fisherman before the rich emperor’
(AECHom I, 38, 578, from: Hogg/Denison 2008: 117)

In Tables 4 and 5 we see the OE demonstratives se/þæt/sēo (from

which the PDE definite article “the”, as well as “that” and “those”
derive) and þes/þis/þēos which do the same job as Modern English
this/these. There is still a significant difference between the use of
se/þæt/sēo and the Modern English article: in the first place, unlike
“the”, demonstratives are (generally) not required, as (2) and (3)
show; secondly, they carry more information – for example, they
11
According to researchers (see, for instance, Baugh/Cable 1993), the whole period
in the history of the English language from 450 to 1150 can be called “Old
English”, since this language was spoken by the Anglo-Saxons from the fifth
century; nevertheless, the first texts in Old English, which represent the only
historical source that philologists examined to study the development of the
language, appeared in the seventh century.
12
The Sermones Catholici, or Homilies of AElfric, Vol.1, ed. Thorpe F.S.A. (London, 1844)
 Nicholas Catasso 19

can express location (Van Gelderen, 2006: 59; 2000: 38):

(2) Þæt hio Beowulfe …|… medoful ætbær

That she-NOM Beowulf-DAT … maedcup at-bore

‘that she brought Beowulf the meadcup’

(Beowulf, vv. 623-2413, from: Van Gelderen,
A History of the English Language: 58)

(3) æþele cempa self mid gesíðum

noble fighter-NOM self-NOM with follower-DAT PL

‘The noble fighter himself with his followers’

(Beowulf, vv. 1312-13, from: Van Gelderen,
A History of the English Language: 59)

Table 4. Demonstratives “the”, “that”, “those”

Masculine Neuter Feminine Plural
Nominative se sēo
þæt Þā
Accusative þone Þā
Genitive þæs þāra, þǣra
þǣre
Dative þām Þām
Instrumental þӯ, þon

Table 5. Demonstratives this”, “these” in Old English

Masculine Neuter Feminine Plural
Nominative þes þēos
þis Þās
Accusative þisne þās
Genitive þisses Þisra
þisse, þisre
Dative þissum Þissum
Instrumental þӯs

13
http://www.humanities.mcmaster.ca/~beowulf/main.html
20 The Grammaticalization of Demonstratives: A Comparative Analysis

Texts in this period show the use of demonstratives as well as the

frequent pre-nominal position of this determiner, especially in Late
Old English (Swanton 1996):

(4) …he hæfde þæt rice

14
‘He held the kingdom’

(Anglo-Saxon Chronicle: Introduction, from:

Michael Swanton, The Anglo-Saxon Chronicles, London)

From ME (particularly from Late Middle English) onwards, the

definite article (and later also a possessive pronoun) is the more
usual option. The two demonstratives indicating number (this-these
and that-those) become very frequent, as well. Both articles and
demonstratives lack case distinctions (Josipovici 1971: 64):

(5) And with his fest he smoot me on the heed

‘And with his fist he hit me on the head’
15
(Chaucer WBProl , v. 801, from: Josipovici, The world
and the book: a study of modern fiction, Stanford)

What follows is the beginning of the General Prologue from the

Canterbury Tales by Geoffrey Chaucer. The text shows that the use
of the definite article has become widespread (translation by Nevill
Coghill):

Whan that Aueryłł wt his When in April the sweet

shoures soote, showers fall

14
http://www.georgetown.edu/departments/medieval/labyrinth/library/oe/texts/asc/a.html
15
See Chaucer, Geoffrey, The Wife of Bath's Prologue (ll.1-862), in The Canterbury
Tales http://www.librarius.com/cantales.htm
 Nicholas Catasso 21

The droghte of Marcħ, hath And pierce the drought of

perced to the roote; March to the root, and all

And bathed euery veyne in The veins are bathed in liquor

swich lycour of such power

Of which vertu engendred is As brings about the

the flour16; engendering of the flower;

Old English demonstratives could function as relatives, which is

no longer the case in Middle English: that becomes the relative
marker, as in the following example, from a slightly later version of
Layamon:

(6) after þan flode. Þat fram God com. Þat al ere acwelde.
after that flood which from God came which all here killed
‘after the flood which came from God (and) which killed
all (creatures) here’
(Layamon’s Otho 7-9)17

Table 6 shows the combinations of the definite article the and the
noun sun in all numbers and cases as in Chaucer (Van Gelderen
2006: 124):

16
http://www.librarius.com/cantales.htm
17
MS. Cott. Otho, C. XIII., taken from Layamon (c. 1215), Madden, Frederic, ed.,
Layamons Brut, or Chronicle of Britain; A Poetical Semi-Saxon Paraphrase of
The Brut of Wace, I, London: The Society of Antiquaries of London, 1847: 2
(also cfr. http://etext.lib.virginia.edu/toc/modeng/public/LayBruO.html)
22 The Grammaticalization of Demonstratives: A Comparative Analysis

Table 6. The Definite Article in Combination with the Noun “Sonne”

in ME
Singular Plural
Nominative the sonne the sonnes
Genitive the sonnes the sonnes
Accusative/Dative the sonne the sonnes

3. The Question of the Origin

of the Category “Article”

As Giusti (1993) – among others – points out, the origin of the

definite article in Germanic languages coincided with the loss of
morphological case. From a chronological point of view, bearing in
mind that grammaticalization processes and generally speaking
diachronic language change do not result in different distinct stages
of a language but in a historical continuum, in late Old English there
was a certain amount of syncretism in the case system of English,
although it was still distinctive. The case system dramatically collapsed
in the transition from OE to ME. As a generally accepted explanation,
the morphological loss of case in English was triggered by
phonological changes (for instance, the reduction of many word-final
vowels to schwa entailed that many distinctions were lost; moreover,
word-final nasal consonants were eliminated). But this phenomenon
has motivated in various ways depending on the theoretical
framework of reference. According to another explanation (Askedal
1999 among others), the loss of morphological case was a part of the
development from synthetic to analytic languages. In the generative
framework this change is explained as a passage from inherent to
structural case (Lightfoot 1999: 132-33). As in Table 2 and 3, Old
English had four cases (nominative, genitive, accusative and dative)
and a vestigial instrumental, which disappeared in the period from
 Nicholas Catasso 23

the 10th to the 13th century, the loss spreading through the population
from the North to the South, probably under the influence of the
Scandinavian settlements (O’Neil 1978). The loss of a case system
generated confusion in the texts as far as case endings are concerned.
Interestingly enough, some remnants of the case system still exist in
English, for example in the use of pronouns (from this point of view,
it will be useful to compare English and German, a language which
still has a definite case system). The same goes for Romance
languages, which have not preserved the Latin system:

(7) I am going to Brighton tomorrow. Would you come with me?

(8) Ich fahre morgen nach Brighton. Kommst du mit mir?

(9) (Io) vado a Brighton domain. Vieni con me?

(10) Je vais à Brighton demain. Tu viens avec moi?

Considering the system of turning NPs into arguments through

morphological case, Osawa argues in his article The emergence of
DP in the history of English: The role of the mysterious genitive
(2007: 135) that genitive lost this function earlier than the rest of
the morphological case system in English: genitive-marked nouns
very rarely occurred as arguments of predicates or adjectives already
in Early Middle English (exception made for specific dialectal
realities such as the Southern dialects where verbs like abide kept
the genitive) and the genitive case came to occur almost exclusively
in noun phrases. Some OE genitive nouns were replaced by “of-
phrases”. Philippi (1997: 65) defines the article in the modern
Germanic languages – referring to its use – as a “default reference
marker” and relates the development of the determiners in the
Germanic languages to the loss of genitive as an object case. But –
following Philippi’s line of inquiry – in order to understand when
24 The Grammaticalization of Demonstratives: A Comparative Analysis

the article developed from demonstratives, it is necessary to

discriminate between them and their functions, as the classification
of an item in texts as either a demonstrative or an article is not
always straightforward.
According to Lyons’ analysis, the definite article bears the
semantic feature [+Definite], whereas demonstratives are [+Definite],
[+/-Proximal] 18 and [+Demonstrative] 19 . To establish the actual
value of the demonstrative/article, syntactic evidence is much more
reliable than mere form: the OED provides examples of “abnormal
uses of se in oblique cases, and of sa plural, ses genitive singular”,
specifying that in some cases “s could be a scribal error for þ”.
Considering, therefore, syntactic rather than formal textual elements,
there are basically two ways to determine the value of an item: (i)
consider the number and case of its complement; (ii) consider
whether or not it takes a complement. That is to say, one marker of
the change is that writers started to introduce nouns other than
masculine singulars with se or þe. Another, possibly even more
important, factor is the occurrence and position of the item in the
sentence. From this point of view, Giusti (2002: 50) points out that

18
The definition [+/-Proximal] indicates the commonly accepted distinction between
proximal demonstratives, referring to objects which are physically close to the
speaker (PDE this), and distal demonstratives (PDE that), indicating objects
further removed from the speaker. Italian, as well as other languages like
Georgian and Spanish traditionally have a third type of demonstrative, the medial
one, which basically refers to an object close to the addressee (e.g. Italian
codesto/codesta).
19
By “definite” we mean here, following Lyons’ fundamental claim (Lyons 1999:
2), the intuitive distinction between cases like “this house” (judged by speakers as
[+Definite], and “several houses” [-Definite]. Demonstratives are conceived in
this framework as definite, but their definiteness is clearly “not a matter of
inclusiveness”. Wood (2005: 169-70) argues that demonstratives are not
necessarily definite, proposing for example that in an utterance as “This man with
long greasy hair and a sleeping bag sort of rolled into a ball comes over and starts
looking in the bins” (BNC A74 2276) the NP “this man” is referential but not
definite.
 Nicholas Catasso 25

the crucial difference between demonstratives and articles is the

impossibility for the latter to appear without an overt sister
projection (i.e. in order for the sentence to be grammatical, the
article should be followed by a noun – further modified, by an
adjective for example, or not), while this is possible with
demonstratives:

(11) I did this

(12) **I did the

Apparent counterevidence to this generalization (Giusti, 2002:

50) can be found not only in German, but also in a number of other
Germanic languages (not in English, as we have seen). I will
consider German here for reasons of convenience:

(13) Ich kenne Peter. Der ist verrückt.

‘I know Peter. He is crazy’

(14) Ich kenne Peter. Er ist verrückt.

‘I know Peter. He is crazy’

(15) Ich kenne den, aber nicht besonders gut.

‘I know him, but not particularly well’

Leiss (2007: 73) notices a significant parallelism between the

(in)definiteness pattern and (im)perfectivity, claiming that the
emergence of the definite article is due to changes in the aspectual
system of a language, in the sense that definiteness/indefiniteness
and perfectivity/imperfectivity are equivalent techniques of nominal
and verbal quantification. She argues that the first occurrences of the
definite article in a language are normally observable in coincidence
with an erosion of the verbal part of the pattern and that there is
26 The Grammaticalization of Demonstratives: A Comparative Analysis

converging evidence from linguistic typology that languages

characterized by an article system tend to avoid aspect and aspect
languages generally avoid article systems.
Interestingly, Hawkins (2004: 83-84) argues that, in principle,
the commonplace of the explanation of the evolution of demonstratives
into articles through the focus on the expanded semantics and
pragmatics of the definite article compared with the demonstrative is
basically an over-simplification: there is no reason why a category
as the definite article should develop out of demonstratives, as it
expresses meanings20 which are perfectly expressible in articleless
languages. Such an expansion of determiners – in particular
languages and in particular historical contexts – is therefore to be
explained looking to the processing of grammar.

4. Implications of the Grammaticalization

of Demonstratives

But, back to the grammaticalization process characterizing the

passage from the distal demonstrative that to the definite article,
interestingly enough Wood (2007: 344) assumes an analysis – in the
generative framework – in which the grammaticalization of the
article involves a lower specifier becoming a higher head. 21 This
means that the process basically consisted in “a loss of the [+Dem]
feature and of the Dem > D movement” (Wood 2007: 344). Following
the minimalistic perspective, she then claims that the definite article
is in PDE the head of DP and is merged there.22

20
The term “meaning” in this context is arguable for many linguists: Giusti (1998,
among others), for instance, considers the definite article to only encode
morphological case and no semantic content.
21
In contrast, for example, to Lyons’ hypothesis (1999: 299), who points out that
determiners such as PDE the are specifiers.
22
Cfr. Van Gelderen (2004), Roberts and Roussou (2003)
 Nicholas Catasso 27

Hopper and Traugott (2003) observe that grammaticalization

involves a loss of concrete semantic content. Very interesting, from
this point of view is Greenberg’s argument-tation on the systematic
development of the article which he conceptualizes as “the cycle of
the definite article” (Greenberg 1978, 1990). Greenberg proposes
three stages through which an originally deictic pronoun (whose
semantics is basically referential) becomes a pure noun marker.23 He
speaks of the common, well-attested origin of the definite article (i.e.
the demonstrative, endowed with deictic force) as the stage zero of
its development. It then loses its deicticity (that is most often one
which points to location near the third person rather than the first or
second person) and acquires an anaphoric value (i.e. it becomes a
definite article), identifying a referent as already mentioned in the
previous discourse (Stage I). The second part of this development
(Stage II) involves a loss of definiteness: here the article introduces
noun phrases with either definite or indefinite specific reference (it
has become a “non-generic article”, according to Greenberg’s
designation). Stage III, in which the item becomes a nominal marker,
is characterized by the compulsory status of the former article in any
NP and is often incorporated to the noun. The general process of
evolution of the demonstrative into the definite article, therefore,
involves both semantic bleaching (i.e. a loss of lexical features) in
that it loses any referential meaning and becomes a pure marker and
a syntactic weakening, in the sense that the D element becomes an
agreement morpheme and consequently undergoes syntactic
decategorization.

23
A noun marker is a generic category of markers which signal that a noun will
follow, often bearing information about its singularity/plurality.
28 The Grammaticalization of Demonstratives: A Comparative Analysis

5. The German Case

In German, the definite article developed from the demonstrative

pronouns ther/der, thiu/diu, thaʒ/daz in the Old High German period.
Giusti (1995: 85) confirms the view, very popular in the philological
literature, according to which the loss of case marking played the
most important role in the emergence of the definite article in
Germanic languages, although the situation in English and in
German is of course different, since English (like Dutch) has lost
morphological case – with clear effects on word order – and German
hasn’t. A comparison between the following two texts (Tschirch
1975: 168; in Giusti 1995: 85) shows that in the 30 year time span
dividing the two translations of the same Latin text by Matthew the
use of the definite article seems to have completed its development.
The first translation is from the Monsee fragment, dated around 800,
the second from Tatian, dated around 830:

(16) a. Mt. 13,1 in illo die (Et) exiens (Jesus) de domo

b. Monsee in demo tage geng Jesus ûz fona hûs
c. Tatian inti ûzgangenti fon themo hûse
in that day (And) going-out of the house
(Jesus)

(17) a. Mt. 13,1 sedebat secus mare

b. Monsee saz bî sêwe
c. Tatian saz nah themo sêwe
was sitting near the see

While in (16) the demonstrative demo directly corresponds to the

demonstrative in the Latin text, in (17) the two instances of themo
do not have a counterpart in Latin and must subsequently be
interpreted as articles. 24 It is to be reminded, however, that the

24
Tschirch (1975) notices, from this point of view: “Mit der Umfunktionierung des
 Nicholas Catasso 29

definite article does not make its appearance out of the blue and
texts do not always show clearly the more or less established
presence in all contexts of this new category; what is more, for
obvious reasons the use of the new category in these stages of the
languages cannot be fully compared to that of Present-Day German.
Let us consider, for instance, the Lord’s Prayer in two Old High
German dialects, Bavarian (early 9th century) and East Franconian
(Tatian, c. 830):

Fater unser, du pist in himilum (Bavarian, Freisinger

Paternoster)
Fater unser, thū thār bist in himile (East Franconian)
Vater unser, der Du bist im Himmel (Modern German)

Here, we see for example that the preposition in remains simple

in both dialects, whereas in Modern German it is made up of the
preposition and the article, as its relation to Himmel expresses
definiteness. This is also the case in the last part of the prayer:

Uzzan kaneri unsih fona allem sunton (Bavarian, Freisinger

Paternoster)
ūzouh arlōsi unsih fon ubile (East Franconian)
sondern erlöse uns von dem Übel (Modern German)25

Oubouzar (1992) notices that the evolution of the category

“demonstrative” – which can be assumed to be in Spec-FP – into

Demonstrativs zum Artikel hat das Ahd. ein ganz einfaches Mittel entwickelt, um
dem Zusammenfall der Kasus zu begegnen: die Aufgabe, die die Kasusendung
einwandfrei nicht mehr zu erfullen vermag, ubernimmt der Demonstrativ, das
damit zum bestimmten Artikel umgeprägt wird – der Aufgabe, die ihm das
grammatische System damit zuweist, hat er sich bis heute gewachsen gezeigt!”
25
Taken from: Braune, Wilhelm, Ebbinghaus, Ernst (1994), Althochdeutsches Lesebuch,
17th edn. Niemeyer, Paris
30 The Grammaticalization of Demonstratives: A Comparative Analysis

the definite article in German can only be considered to be complete

when the semantic relation between the determiner and N’ is
reinterpreted as a morphological relation26.
In Middle High German and Early New High German the use of
the article spreads to usage in forms which are quite similar to those
of Present-Day German already from the early 1100’s:

Table 7. The Definite Article in Middle High German

Masculine Neuter Feminine Plural
Nominative dër daƷ diu die/diu
Genitive dës dës dër dër
Dative dëm dëm dër dën
Accusative dën daƷ die die/diu
Instrumental diu

Nübling (2005: 107) proposes another analysis of the grammati-

calization path taking into account that the German definite article
undergoes a further grammaticalization stage which is usually left
behind by the research, namely the development into prepositional
enclitics (im, zum, am, ans, etc.). This use of the definite article is
not casual: on the one hand, not all article forms can be cliticised
(for instance, die) and not all prepositions can be the basis for this
kind of cliticization (gegenüber, trotz, etc.); on the other hand, there
are contexts in which the position of the article does not depend on
factors such as grammatical economy (a very significant element in
grammaticalization paths), origin of the speaker, style, etc, but must
obligatorily be attached to the preposition to avoid ungrammaticality:
die Entwicklung vom (*von dem) Demonstrativ zum (*zu dem)

26
In fact, the rise of the article in Old High German also had syntactic consequences,
such as the postposition of the genitive article in contrast with its previous
position before the noun.
 Nicholas Catasso 31

Artikel, sie wandert im (*in dem) Gebirge, and so on. The analysis
she proposes, based on Himmelmann’s view (1997: 23, based, in
turn, on Greenberg 1978 and Lehmann 1982), can be summarized as
follows:

Deictic particle + categorical noun > demonstrative pronoun

> demonstrative determiner > weakly demonstrative definite
determiner > definite article > affixal article > noun marker

6. The Development of the Definite Article

in Romance Languages

As regards Romance languages, the rise of the definite article,

category shared by the whole language family, has often been linked
to the loss of the Latin case system. In fact, the presence/absence of
the definite article is one of the most outstanding differences
between Latin and Romance and “Latin grammarians frequently
remarked on the lack of an equivalent of the Greek article in their
own language” (Posner 1966: 126). Therefore, argues Posner, one of
the suggestions which have been elaborated consists of the rise of
the definite article in Christian Latin as a transfer from Greek,
although early Latin translations of the Greek New Testament do not
show any relevant consistency of the representation of the Greek
article in Latin. Faingold (1993: 6) claims that in language history,
as well as in creolization and other emergent systems, definite
articles are not borrowed from the superstrate languages but are
created anew from demonstratives. As in English and German, the
definite article in Romance grammaticalized from a distal pronoun,
namely Latin ILLE, which also constitutes the base of the distal
demonstrative, reinforced by a preposed ECCE (= Behold!). The
Classical Latin demonstratives were increasingly used as articles in
Vulgar Latin from ca. 380 to 1150, indicating a change of function
32 The Grammaticalization of Demonstratives: A Comparative Analysis

of this grammatical category. The initially anaphoric article derived

from a deictic pronoun indicating the spatial position of the referent,
widening then its function to denote: 1. unique nouns; 2. abstracts; 3.
inalienable possession, becoming therefore more specific in
semantic content and syntactic form and undergoing a shift from a
“pointing” to a more expressive and subjective meaning. This kind
of semantic shift is generally considered to be a striking characteristic
of the early stages of grammaticalization: the demonstrative is
initially used exclusively to attract the hearer’s attention and assumes
then a subjective, “affective” use. Interestingly, the demonstrative
deictic reduction was the starting point of the grammaticalization of
both object clitics and the definite article27. According to Epstein
(1994), who considers in particular the case of French, the definite
article in this language developed from expressive uses of the Latin
demonstrative ILLE.
As insisted on by Vincent (1997: 154), the competition between
Latin ILLE and IPSE (i.e. between two demonstratives bearing
mainly anaphoric/distal and contrastive features, respectively) in the
early grammaticalization stage in Late Latin is at the origin of the
role played by ILLE as the item which actually grammaticalized.
According to Selig (1992: 165), it was eventually ILLE that was
grammaticalized because of contextual reasons: it could be used in
more diverse contexts as compared to IPSE, whose meaning was
limited to anaphoric uses. Considering the corpus she analyzed, she
comes to the following conclusions: as regards ILLE, “Der Bereich
der definiten Erstnennung ist die eigentliche Funktionsdomäne von
ILLE in den analysierten Texten”. For IPSE, on the other hand,
“IPSE wird in fast allen hier analysierten Texten in erster Linie zur
Kennzeichnung anaphorisch definiter NPs verwendet” (Selig, 1992:
153). Type frequency and generalization would be then the reasons

27
The double evolution of the deictic seems to be due to the context: ill- with a verb
yields a pronoun, ill- with a noun an article.
 Nicholas Catasso 33

why ILLE, developed into the definite article in French. In Old

French, the masculine nominative article is etymologically related to
the corresponding pronoun, although they are not formally identical,
while in Present-Day French, where the definite article, just as in the
other Romance languages, serves to distinguish between definite
and indefinite NPs, the article forms are identical to the clitic object
pronouns:

ILLĪ > pronoun il, article li

Modern French clitic pronouns/articles: le, la, les
(Les filles sont allées jouer. Regarde-les! )

The Spanish forms el, la, los, las correspond to Vulgar Latin
ILLE, ILLA which in turn derived from Classical Latin ILLE, ILLA,
forms which consistently appear before the noun phrase. Interestingly
enough, the Portuguese forms o, a, os, as correspond to a Vulgar
Latin innovation derived from ILLE, ILLA, i.e. the article variant ea,
eo, which seem to have developed from the accusative forms illa,
illum and still do not display a completely separate semantic value
article-demonstrative even though they often convey the same form,
as we see in the following examples (Faingold 1993: 8, citing
Bernard 1971: 37; 102):

(16) In ea ergo die et in ea hora, qua auertarent Persae aquam

‘On the day and in the hour when the Persians diverted
28
the water’ (The Pilgrimage of Etheria , from: Faingold
1993: 8)

Faingold also argues that this grammaticalization path can be

explained in terms of discourse factors. Comparing Vulgar Latin

28
Itinerarium Aetheriae, tr. by Clara di Zoppola, in: Attilio Agnoletto, Storia del
cristianesimo, IPL, Milano, 1978.
34 The Grammaticalization of Demonstratives: A Comparative Analysis

texts over different centuries, we can mark a diverse pragmatic use

of these items: texts from the 4th-6th century show that Latin
demonstratives are used to refer to noun phrases corresponding to
objects which are prominent in the discourse; texts from the 8th
century already show a significant change in the use of demonstratives
as they are also used to refer to noun phrases that are accessible to
the speaker and the hearer; the further development, from the 12th
century to the modern usage in the Romance languages, shows that
the definite article precedes those noun phrases which can simply be
identified by the speaker and the hearer, regardless of whether they
actually play a prominent role in the discourse.
Rumanian is the only modern Romance language displaying the
use of an enclitic definite article (whilst the indefinite article appears
in front of the noun phrase, just as in Swedish). In this language the
enclitic article is clearly related to the inflected disjunctive pronoun
(Posner 1966: 127):

Fem. mama (‘the mother’ = mǎma+ea)

Masc. fiul (‘the son’ = fiu+el)

A significant syntactic difference in the use of articles-

demonstrative pronouns which these languages have developed in
their history (and still showing the Latin heritage) consists of the
fact that in cases like the following in French an article can be used
with the function of a demonstrative, which is not the case in Italian
and in Rumanian, which needs both the article and the pronoun, as
noted by Posner:

(17) Le livre blanc et le rouge (‘The white book and the red
one’)

(18) … e quello rosso

 Nicholas Catasso 35

(19)… cel roṣul

Faingold (1993) identifies three main types of criteria as central

for recognizing Romance definite articles as a result of the
grammaticalization of Latin demonstratives: statistical criteria,
which are concerned both with a synchronic and a diachronic
increase in text frequency of the emergent article; structural criteria,
for example the analysis of phonological weakening (phonetic
shortening, e.g. Lat. illa > Sp. la and loss of stress) and location,
including pronominal and postnominal position of the Latin
demonstrative (for instance, illa aqua ‘the water’ vs. epistolam
ipsam ‘the letter’); functional criteria, regarding the status of the
NPs being preceded by a demonstrative (nouns which are prominent
in the narration or which are “accessible” due to prior mention).

7. The Absence of the Definite Article

in Slavic Languages

As is well-known, Baltic and Slavic languages (except for

Bulgarian and Macedonian) lack a definite article. As we have seen,
the rise of the article is often linked to the loss of morphological
case marking which many languages underwent in different periods
of their history. Even if in some languages articles are found in
coincidence with the presence of case declinations, a correlation is
definitely present between these two developments29, as Bulgarian
and Macedonian are the only two languages in the family which

29
From this point of view, it is interesting to consider Renzi’s explanation (1992) of
the logical order in which the two phenomena took place: 1. The development of
the article, marked by case; 2. Loss of nominal inflections made possible by the
new affix. According to Renzi’s hypothesis, this was the order that languages like
Bulgarian and Macedonian followed in their diachronic development, as they lost
morphological case.
36 The Grammaticalization of Demonstratives: A Comparative Analysis

have developed the article category, which is postponed or

agglutinated, after the loss of morphological case. The case of
Bulgarian and Macedonian represents a further confirmation of the
development of the definite article from demonstratives, since this
category in the two languages evolved out of the Old Slavic
demonstrative pronoun *jb, *ja, *je (‘this’), a form which
disappeared from the spoken language by the 16th century but was
maintained in the literary language (Gebert 1996: 11). Gebert also
claims that a striking characteristic of the Slavic languages which
have not developed the definite article is the presence, attested in
Old Slavic, of long and short forms of the adjective, which linguists
generally define as definite and indefinite30 , respectively, and are
today still productive in Slavic languages. Long forms of the
adjective are interestingly composed of the adjective and the
demonstrative pronoun *jb, *ja, *je, that, in turn, developed from
the Indo-European deictic *-io and belongs to the same class as the
Latin demonstrative is.
Heine and Kuteva (2003, 2005) propose that grammaticalization
and contact-induced language change are not mutually exclusive and
they may jointly conspire in triggering grammatical change.
Talking about areal convergence in grammaticalization processes,
Giacalone Ramat (2008) draws on data from a recent study on the
use of demonstrative pronouns in Czech, Sorbian and Slovenian
(Trovesi 2004) to hypothesize that this category is undergoing a
grammaticalization path in these languages. Giacalone Ramat and
Trovesi’s interesting assumption, diverging from – or at least
offering new stimuli to – the classical theory on the origin of the
article in Indo-European languages is that this development of ten,
tón, ta is taking place under the influence of German, since the
German language played in history a significant cultural and
political role in the areas where Czech, Sorbian and Slovenian are

30
This definition depends on the fact that they express definiteness and indefiniteness.
 Nicholas Catasso 37

spoken. Heine and Kuteva’s claim that the Sorbians simply

replicated the article category having as a starting point the German
article seems to be inadequate in that the category has not yet fully
arisen in Sorbian and in the other Slavic languages in question, as
the grammaticalization is still at an incipient stage; it is now arising
in the mind of bilingual individuals.

8. Conclusion

In this paper I examined the concept of grammaticalization in all

its facets (from a general to a more author-specific definition, from
its implications in pragmatics to its role in the generative framework
and to the studies which call into question the validity of the
grammaticalization theory, etc.). I also analyzed the particular case
of the grammaticalization of demonstrative pronouns into the
definite article in crosslinguistc and diachronic perspective, focusing
on English and German as representatives of the Germanic language
family and mentioning the development or absence of this category
in Romance and Slavic languages. The reasons generally linked to
this kind of grammaticalization path are linked to the loss of
morphological case marking in a certain stage of the development of
the language. In the course of this development, demonstratives lose
their deictic function and turn into formal markers of definiteness.
Among the Indo-European languages, the grammatical category of
the definite article is a common characteristic in the Germanic and
in the Romance family, whereas Baltic and Slavic languages lack a
definite article. The only exceptions in the latter case are
Macedonian and Bulgarian, which developed a post-posed definite
article after the loss of the morphological case system. There are
also cases, in particular among the Slavic languages (Sorbian, Czech,
Slovenian), which can be considered to be under way in their
grammaticalization path demonstrative > definite article.
38 The Grammaticalization of Demonstratives: A Comparative Analysis

References

Aitchison, J. 2001. Language Change: Progress Or Decay? 3rd ed.

Cambridge: Cambridge University Press.
Alexiadou, A., L. Haegeman & M. Stavrou. 2007. Noun Phrase in
the Generative Perspective (Studies in Generative Grammar
71). Berlin - New York: Mouton de Gruyter.
Allan, K. 2001. Natural Language Semantics. Oxford & Malden:
Blackwell Publishers.
Althaus P., H. Henne & H.E. Wiegand. 1973. Lexikon der
germanistischen Linguistik, Tübingen: Niemayer.
Anderson, S. R. 2005. Aspects of the Theory of Clitics Oxford:
Oxford University Press.
Anttila, R. 1972. An Introduction to Historical and Comparative
Linguistics. New York: McMillan.
Askedal, J. O. 2001. ‘Oblique Subjects’, Structural and Lexical Case
Marking: Some Thoughts on Case Assignment in North
Germanic and German. Faarlund, J. T. (ed.) Grammatical
Relations in Change, 65-97.
Baugh, A. C. & T. Cable. 1993. A History of the English Language.
London: Routledge.
Beekes, R. 1995. Comparative Indo-European Linguistics: an
Introduction. Amsterdam: John Benjamins.
Bianchi, V. 1999. Consequences of Antisymmetry: Headed Relative
Clauses, Berlin New York: Mouton de Gruyter.
Bisang, W., N. Himmelmann & B. Wiemer. 2004. What Makes
Grammaticalization? A Look from its Fringes and its
Components. Berlin New York: Mouton de Gruyter.
Booij, G., C. Lehmann & J. Mugdan. 2000. Morphologie. Ein
internationales Handbuch zur Flexion und Wortbildung. Berlin
- New York: Mouton de Gruyter.
Braune, W. & E. Ebbinghaus. 1994. Althochdeutsches Lesebuch,
17th ed. Paris: Niemeyer.
 Nicholas Catasso 39

Brinton, L. J. & E. Traugott. 1995. Lexicalization and Language

Change. Cambridge: Cambridge University Press.
Campbell, L. 1998. Historical Linguistics: An Introduction. Cambridge,
MA: MIT Press.
. 2001. What's Wrong with Grammaticalization? Language
Sciences 23, 113-161.
Chambers, J. K. 1995. Sociolinguistic Theory: Linguistic Variation
and its Social Significance, Oxford: Blackwell.
Cinque, G. & G. Giusti. 1995. Advances in Roumanian Linguistics.
Amsterdam: John Benjamins.
Crespo, E., J. De La Villa & A. R. Revuelta. 2006. Word Classes
and Related Topics in Ancient Greek: Proceedings of the
Conference on ‘Greek Syntax and Word Classes’ Held in
Madrid June 2003, Madrid: Peeters Publishers.
Cruttenden, A. & D. Faber. 1991. The Accentuation of Prepositions.
Journal of Pragmatics 15, 265-286.
Crystal, D. 1980. A First Dictionary of Linguistics and Phonetics.
Boulder, CO: Westview Press.
Diessel, H. 1999. Demonstratives: Form, Function, and Gramma-
ticalization. Amsterdam Philadelphia: John Benjamins.
. 1999. The Morphosyntax of Demonstratives in Synchrony and
Diachrony. Linguistic Typology 3, 1-49.
Diessel, H. 2003. The Relationship Between Demonstratives and
Interrogatives. Studies in Language 27, 581-602.
Diewald, G. 2006. Context Types in Grammaticalization as Cons-
tructions. Constructions 1-9, 2006. Available at URL
<http://www.constructions-online.de/articles/specvol1/>
Eckardt, R. 2008. Meaning Change in Grammaticalization: An
Enquiry into Semantic Reanalysis. Oxford: Oxford University
Press.
Einhauser, E. 1989. Die Junggrammatiker. Ein Problem für die
Sprachwissenschaftsgeschichtsschreibung. Trier: Wissenschaftlicher
Verlag.
40 The Grammaticalization of Demonstratives: A Comparative Analysis

Epstein, R. 2001. The Meaning of Definite Articles in Cross-

Linguistic Perspective. In: Eniko Németh (ed.). Cognition in
Language Use: Selected Papers from the 7th International
Pragmatics Conference, Antwerp: International Pragmatics
Association, 174-189.
Faingold, E. 1993. The Development of the Definite Article from
Latin to Spanish and Portuguese. Paper presented at the 1993
Annual Meeting of the Linguistic Society of America, Los
Angeles, California. Jan. 1993.
Fillmore, C.J. 1988. The Mechanism of ‘Construction Grammar’,
Proceedings of the Fourteenth Annual Meeting of the Berkley
Linguistic Society. Berkley: University of California Press. 35-
55.
Fischer, O, M. Norde & H. Perridon. 2004. Up and down the Cline -
The Nature of Grammaticalization. Amsterdam: John Benjamins.
Fortson, B. W. 2004. Indo-European Language and Culture.
London: Blackwell Publishers.
Fox, A. 1995. Linguistic Reconstruction. An Introduction to Theory
and Method. Oxford: Oxford University Press.
Gebert, L. 1996. Riflessioni sull'Articolo Mai Nato nelle Lingue
Slave. In Bernacchio R. et al., Determinatezza e Indeterminatezza
nelle Lingue Slave, Padova: Unipress.
Ghomeshi, J., I. Paul & M. Wiltschko. 2009. Determiners:
Universals and Variation. Amsterdam: John Benjamins.
Giacalone R., A. 2008. Areal Convergence in Grammaticalization
Processes. In López-Couso, M. J. & E. Seoane (eds.),
Rethinking grammaticalization: New Perspectives for the Twenty-
First Century. Amsterdam - Philadelphia: John Benjamins.
Giusti, G. 1993. La Sintassi dei Determinanti. Padova: Unipress.
Giusti, G. 1999. The Functional Structure of Noun Phrases. A Bare
Phrase Structure Approach. Working Papers in Linguistics 9,
105-160.
Givón, T. 1979. On Understanding Grammar. New York: Academic
 Nicholas Catasso 41

Press.
Givón, T. 1991. Markedness in Grammar: Distributional, Communi-
cative and Cognitive Correlates of Syntactic Structure. Studies
in Language 15, 335-370.
Grice, H. P. 1975. Logic and Conversation. In Cole, P. & J. Morgan,
Syntax and Semantics. Speech Acts. New York: Academic Press.
Greenberg, J. H. 1990. How Does a Language Acquire Gender
Markers? In Denning, K. & S. Kemmer. On Language:
Selected Writings of Joseph H. Greenberg. Stanford: Stanford
University Press.
Greenberg, J. H. 1991. The Last Stages of Grammatical Elements:
Contractive and Expansive Desemanticization, in Traugott and
Heine, eds., Vol.1, Approaches to Grammaticalization,
Amsterdam: John Benjamins.
Haegeman, L. 1997. The New Comparative Syntax. London:
Longman.
Haider, H., S. Olsen & S. Vikner. 1995. Studies in Comparative
Syntax. Dordrecht: Kluwer Academic Publishers.
Halliday, M. 1961. Categories of the Theory of Grammar in
Halliday(ed.) System and Function in Language. Oxford: Kress,
52-72.
Hammond, M. & M. Noonan. 1988. Theoretical Morphology:
Approaches in Modern Linguistics, San Diego: Academic Press.
Harris, A. C. 2002. Endoclitics and the Origins of Udi Morphosyntax.
Oxford: Oxford University Press.
Haspelmath, M. 1998. Does Grammaticalization Need Reanalysis?
Studies in Language 22, 315-372.
Haspelmath, M. 1999. Why Is Grammaticalization Irreversible?
Linguistics 37, 1043-1068.
Hawkins, J. A. 2004. Efficiency and Complexity in Grammars.
Oxford: Oxford University Press.
Heine, B. & M. Reh 1984. Grammaticalization and Reananalysis in
African Languages. Hamburg: Buske.
42 The Grammaticalization of Demonstratives: A Comparative Analysis

Heine, B. & T. Kuteva. 2002. World Lexicon of Grammaticalization.

Cambridge: Cambridge University Press.
Hopper, P. J. & E. Traugott. 2003. Grammaticalization, 2nd ed.
Cambridge: Cambridge University Press.
Heine, B. & T. Kuteva. 2003. On Contact-induced Grammaticali-
zation. Studies in Language 27, 529-272.
Heine, B. & T. Kuteva. 2005. Language Contact and Grammatical
Change. Cambridge: Cambridge University Press.
Himmelmann, N. P. 1997. Deiktikon, Artikel, Nominalphrase: Zur
Emergenz syntaktischer Struktur. Tübingen: Niemeyer.
Hogg, R. & D. Denison. 2008. A History of the English Language.
Cambridge: Cambridge University Press.
Hopper, P. & E. Traugott. 2003. Grammaticalization, 2nd ed.
Cambridge: Cambridge University Press.
Humboldt, W. 1822. On the Origin of Grammatical forms and their
Influence on the Development of Ideas. In: P. Lang. 1997.
Essays on Language. Frankfurt am Main: Europäischer Verlag
der Wissenschaften.
Jeffers, R. J. & A. M. Zwicky. 1980. The Evolution of Clitics. In
1980. Papers from the 4th International Conference on
Historical Linguistics. E. Traugott, R. LaBrum & S. Shephard.
(eds.) Amsterdam: John Benjamins.
Joseph, B. 2001. Is There Such a Thing as “Grammaticalization”?
Language Sciences 23, 163-186.
Josipovici, G. 1971. The World and the Book: A Study of Modern
Fiction. Stanford: Stanford University Press.
Jucker, A. H., G. Fritz & F. Lebsanft. 1999. Historical Dialogue
Analysis. Amsterdam: John Benjamins.
Juvonen, P. 2000. Grammaticalizing the Article – A Study of
Definite Adnominal Determiners in a Genre of Spoken Finnish.
Stockholm: Stockholm University.
Krug, M. G. 2000. Emerging English Modals. A Corpus-based
Study of Grammaticalization. New York & Berlin: Mouton de
 Nicholas Catasso 43

Gruyter.
Kuriłovicz, J. 1975. The Evolution of Grammatical Categories.
Esquisses Linguistiques II, 38-54.
Lane, G. S. 1961. On the Formation of the Indo-European
Demonstrative. Language 37, 469-475.
Lass, R. 1997. Historical Linguistics and Language Change.
Cambridge: Cambridge University Press.
Leuschner, T., T. Mortelmans & S. De Groodt. 2005.
Grammatikalisierung im Deutschen, Berlin: Mouton De
Gruyter.
Lightfoot, D. 1999. The Development of Language: Acquisition,
Change and Evolution. Oxford: Blackwell.
Lyons, C. 1999. Definiteness. Cambridge: Cambridge University
Press.
Marouzeau, J. 1953. Lexique de la Terminologie Linguistique. Paris:
Geuthner.
McMahon, A. 1994. Understanding Language Change. Cambridge:
Cambridge University Press.
Meillet, A. 1912, L’Evolution des Formes Grammaticales. Scientia
12, 384-400.
Müller, H. H. & A. Klinge 2008. Essays on Nominal Determination.
Amsterdam Philadelphia: John Benjamins.
O’Neil, W. 1978. The Evolution of the Germanic Inflectional
System: A Study of the Causes of Language Change. Orbis 27,
246-86.
Oniga, R. 2004. Il Latino. Breve introduzione linguistica. Milano:
Franco Angeli.
Osawa, F. 2007. The Emergence of DP in the History of English:
The Role of the Mysterious Genitive. In Historical Linguistics
2007. Selected Papers from the 18th International Conference
on Historical Linguistics, M. Dufresne, F. Dupuis & E. Vocaj.
(eds.) Montréal, Québec, August 2007. Amsterdam: John
Benjamins.
44 The Grammaticalization of Demonstratives: A Comparative Analysis

Oubouzar, E. 1992. Zur Ausbildung des bestimmten Artikels im

Althochdeutschen. In Y. Desportes (ed.) Althochdeutsch.
Syntax und Semantik. Lyon, 71-87.
Parotté, W. & R. Dirven. 1985. The Ubiquity of Metaphor.
Amesterdam Studies in the Theory and History of Linguistic
Science IV. Amsterdam: John Benjamins.
Philippi, J. 1997. The Rise of the Article in the Germanic Languages.
In Parameters of Morphosyntactic Change, Van Kamenade, A.
& N. Vincent. (eds.) Cambridge: Cambridge University Press,
62-93.
Posner, R. 1966. The Romance Languages. A Linguistic Introduction.
New York: Doubleday.
Radden, G. & K. U. Panther 2004. Studies in Linguistic Motivation.
Cognitive Linguistics Research. Berlin New York: Mouton de
Gruyter.
Ramat, P. 1987. Linguistic Typology. Berlin: Mouton de Gruyter.
Renzi, L. 1992. Le Développement de l'article en Roman. Revue
Roumaine de Linguistique 2, 161-176.
Ringe, D. 2006. A Linguistic History of English. Volume I. From
Proto-Indo-European to Proto-Germanic Oxford: Oxford
University Press.
Rizzi, L. 1997. The Fine Structure of the Left Periphery. In: L.
Haegemann (ed.), Elements of grammar. Dordrecht: Kluwer,
281-337.
Roberts, I. 1993. A Formal Account of Grammaticalization in the
History of Romance Futures. Folia Linguistica Historica 13,
219-258.
Roberts, I. & A. Roussou. 2003. Syntactic Change, A Minimalist
Approach to Grammaticalization. Cambridge: Cambridge
University Press.
Salmons, J. C. & S. Dubenion-Smith 2007. Historical Linguistics
2005: Selected Papers from the 17th International Conference
on Historical Linguistics, Amsterdam Philadelphia: John
 Nicholas Catasso 45

Benjamins.
Selig, M. 1992. Die Entwicklung der Nominaldeterminanten im
Spätlatein. Script-Oralia 26, 238-260.
Sonderegger, S. 1979. Grundzüge Deutscher Sprachgeschichte.
Diachronie des Sprachsystems. Bd. 1: Einführung, Genealogie,
Konstanten. Berlin New York: Mouton de Gruyter.
Song, J. J. 2001. Linguistic Typology: Morphology and Syntax.
Harlow and London: Longman.
Stark, E., E. Leiss & W. Abraham. 2007. Nominal Determination:
Typology, Context Constraints, and Historical Emergence.
Amsterdam: John Benjamins.
Swanton, M. 1996. The Anglo-Saxon Chronicles. London: J.M. Dent.
Traugott, E. 1982. From Propositional to Textual to Expressive
Meanings: Some Semantic-Pragmatic Aspects of Grammati-
calization. In W.P. Lehmann & Y. Malkiel. Perspectives on
Historical Linguistics. Amsterdam: John Benjamins.
Traugott, E. 1995. Subjectification in Grammaticalisation. In: D.
Stein and S. Wright, Subjectivity and Subjectivisation.
Cambridge: Cambridge University Press.
Traugott, E. & R. Dasher. 2002. Regularity in Semantic Change.
Cambridge: Cambridge University Press.
Trovesi, A. 2004. La genesi di articoli determinativi. Modalità di
espressione della definitezza in ceco, serbo-lusaziano e sloveno.
Pavia: Franco Angeli.
Tschirch, F. 1975. Geschichte der Deutschen Sprache. Berlin: Erich
Schmidt Verlag.
Van Gelderen, E. 2000. A History of English Reflexive Pronouns:
Person, Self, and Interpretability. Amsterdam: John Benjamins.
Van Gelderen, E. 2004. Grammaticalization as Economy.
Amsterdam: John Benjamins.
Van Gelderen, E. 2006. A History of the English Language.
Amsterdam - Philadelphia: John Benjamins.
Van Kamenade, A. & N. Vincent. 1997. Parameters of Morpho-
46 The Grammaticalization of Demonstratives: A Comparative Analysis

syntactic Change. Cambridge: Cambridge University Press.

Van Kemenade, A. 2000. Jespersen's Cycle Revisited: Formal
Properties of Grammaticalization. In S. Pintzuk al. Diachronic
Syntax. Models and Mechanisms Oxford: Oxford University
Press.
Van Kemenade, A. & B. Los. 2006. The Handbook of the History of
English. London: Blackwell.
Von der Gabelentz, G. 1901. Die Sprachwissenschaft. Ihre Aufgaben,
Methoden und bisherigen Ergebnisse. Leipzig: Tauchnitz.
Wanner, D. 1987. The Development of Romance Clitic Pronouns:
From Latin to Old Romance. Berlin: Mouton de Gruyter.
Wood, J. 2007. Demonstratives and Possessives: From Old English
to Present-Day English. In E. Stark, E. Leiss, A. Werner (eds.),
Nominal Determination:Typology, Context Constraints, and
Historical Emergence. Amsterdam: John Benjamins.